International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 06 | June-2016 www.irjet.net p-ISSN: 2395-0072
© 2016, IRJET ISO 9001:2008 Certified Journal Page 2198
Opinion Mining of Customer Reviews based on their Score using
Machine Learning Techniques
Miss. Lovenika Kushwaha
1
, Mr. Sunil Damodar Rathod
2
1
PG Student, Computer Engineering Department, Dr. D.Y.Patil SOE Lohegaon, SPP University Pune, India
2
Asst. Prof., Computer Engineering Department, Dr. D.Y.Patil SOE Lohegaon, SPP University Pune, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Websites for online shopping is becoming more
and more popular nowadays. Companies are eager to know
about their customer buying behavior to increase their
product sale. Extracting knowledge from large database, Data
Mining is the key approach to use for accurate result. But in
our context, we have to process customer reviews from large
E-commerce, database for which Opinion Mining is the best
approach for mining customer reviews about the product. The
widely available internet resources are letting the users to
shop any products anywhere, anytime at any cost. With the
brisk development in the 3G and 4G we can expect a
tremendous development in the area of M-commerce and
E-commerce. In existing papers, opinion mining is used to
process the online product reviews, feature and recommend
the best product among others. Natural Language Processing
(NLP) and Naive Bayes classification both are used to
determine the polarity of reviews (obtain a polarity score from
negative review and positive review). In this paper a novel
technique is proposed for opinion mining and feature
extraction of product reviews. The objective is to encourage
the customers and assist them in choosing the right product. It
is based on natural language processing, opinion mining and
AdaBoost classifier. Results indicate that the proposed
methods are highly effective and efficient in performing their
tasks. We will also aim at improving the accuracy of our
opinion polarity detection and feature extraction among other
techniques.
Key Words: Opinion Mining, Part-of-speech (POS)
Tagging, Natural Language Processing (NLP), Sentiment
Analysis, Naïve Bayes Classification and AdaBoost
classifier.
1. INTRODUCTION
As we all know very well that E-Commerce sites are gaining
popularity across all over the world. Customers are
migrating towards online purchases more instead of going to
the markets because of its easiness, convenience, reliability,
and rapidness. There are a number of Online shopping
websites that are available on the internet, such as Amazon,
Flipkart, Snapdeal, Jabong, Myntra, Paytm, Zovi, etc. These
websites allow the users to buy products with ease and
lesser prize. A lot of attractive and day-to-day useful
products like books, electronic goods, home appliances,
clothing, and footwear are sold from these sites. These
websites provide an option to the customers to write their
review about their product that they buy from these sites.
These reviews or opinions are very helpful to the users,
manufacturers of the product as well as the developers of the
website. The users who are in quandary to buy a product can
read the reviews about the particular product from these
websites so that they can have a view about their product
before buying it and also know which is on the 1
st
position.
Potential buyers can make decisions based on the reviews of
customers who have purchased and experienced the
product. The manufacturers of the product will be able to
know the minor or major drawbacks of the product from the
reviews which helps the manufacturers to get a chance to
release the updated version of the product which satisfies
the reviews that are mentioned in the websites. Hence online
reviews play a significant role in understanding the